Week 7: July 24 - July 28
After melting the metal out of the silicone model, I was asked to help out with a different project which entailed me working with other members of the lab solely at WIMR. The project was based around getting 4D MRI data of patient-specific TCPC models, and the work I mainly helped with differed considerably from what I had been doing with Rafael.
50% Stenosis Model
I spent most of Monday trying to melt all of the metal out of the finished model. Although I was able to melt most of the core out by submerging the model in boiling water, a metal residue was left on the silicone that seemed to be stuck to it. The only way to get it out was to scrub it out with alcohol. This proved difficult as the only way to reach the metal coating inside the one centimeter-wide opening of the model was using a glass stir rod with a paper towel taped to it. Eventually, however, I managed to completely clean out all of the metal. The finished product was a 50% stenosis silicone model that was transparent and smooth on the inside. Rafael, Luis, and Armando were really impressed with how it had come out. In fact, they thought it was overall better than the previous one that was made. However, the process I used had quite a few limitations. For one, although it was less likely to fail, the time it took from start to finish was a few days longer than the process Luis and Armando had used, and it required more silicone and 3D-printing resin. Moreover, the method I used for producing the metal core would be difficult to apply to more complex models, such as an aneurysm. Upon reflection, the process would probably benefit more if the 3D-printed molds for producing metal cores were redesigned instead of avoided. Nonetheless, the process I used had produced a good model that could consistently be achieved.
Measuring Inlet/Outlet Flow for TCPC Models
Total Cavopulmonary Connection (TCPC) is a surgical process where the heart is reconfigured so that the right ventricle is bypassed. TCPC is necessary in patients whose right ventricles do not sufficiently pump blood to the lungs. The CVFD lab was interested in analyzing several patient specific TCPCs to see what such a procedure would entail for the patient long term. Before taking 4D MRI data, David, another grad student in the lab, wanted me to take data of the inlet/outlet flow rates of the 3D-printed patient-specific TCPC models they had in order to obtain flow rates for CFD (computational fluid dynamics) and confirm data they would later collect. For this, I hooked the models up to the perfusion pump, ran it at several settings, and used a flow probe to measure the volumetric flow rate at each inlet and outlet. I then recorded the data in my notebook and later transferred it to an Excel spreadsheet.
While seemingly straightforward, this process involved constant adjustments and rechecking of data. For example, if the volumetric flow rate of the main inlet for the whole TCPC model was 1.00 L/min, I would expect the sum of the flow rates of the sub-inlets to be 1.00 L/min; however, there were multiple times when this was not the case, and I would have to inspect the model to see what was happening. Sometimes, the issue was that there was a leak in the model; other times, the connectors for the tubing were disturbing the flow at the point I was measuring. All in all, this was a good example for me of how real-world science differed from what I had experienced in a classroom environment. In a school lab, I can usually expect to collect data that seems reasonable, and if it doesn't, it is usually obvious what the problem is. On the other hand, in this experiment, I was never surprised when I collected data that seemed strange, even though the issue which had resulted in such data was rarely obvious.